All of AllAmericanBreakfast's Comments + Replies


It’s almost impossible to predict the future. But it’s also unnecessary, because most people are living in the past. All you have to do is see the present before everyone else does.

That really is a nice quote, though.

4gianlucatruda4dWas about to comment the same thing. Saving it to my Wisdom List.
Self-study ideas for micro-projects in "abstract" subjects?

You're just gonna have to be the sort of person for whom solving the wave equation of the hydrogen atom is a juicy project.

That's a nice quote. I'm working on a handbook for how to self-study. Mind if I (tentatively) use it?

4Charlie Steiner4dGo for it - though it's about selection effects as much as it's advice.
Five Suggestions For Rationality Research and Development

6. Treat the problem of measurement with the same care you treat the intervention. Much of rationality training will have its origins in introspection and intuitive inference. Figuring out what effects your hypothesis ought to predict, precisely enough to come up with a solid form of measurement, might be a lot harder than designing an intervention. Just because you don't have a good way to measure what you're doing doesn't mean it has "no evidence." But if you can find a way to measure, that can be very powerful.

4ChristianKl4dIt's worth noting that inventing good ways of measuring issues is as important for developing the field as developing interventions. I'd love to see posts on LessWrong that purely focus on the way to measure a particular issue inside of rationality. We had some of that in the past with credence calibration but it would be great if talking about how to measure what we care about would a larger part of the rationalist discourse.
Experiments with a random clock

This is a great idea. I find it not impossible that if you build an arduino watch implementing this, that you could make a nice little business patenting (?) and selling them. I'm sure many people would say "why not just learn to enjoy being a little early?" But for whatever reason, some people just don't have much success with that. A gadget that corrected their problem might do just the trick.

Self-study ideas for micro-projects in "abstract" subjects?

Many of the coolest and most useful activities for learning are sealed off from non-professionals, or at least are expensive and time confusing to obtain certification or access. Usually for good reason.

This seems like a fundamental dilemma of the role of school. To make students directly see what’s cool about different subjects, they need lots of hands-on experience. But the vast majority of their time, and most of their evaluation prior to grad school, comes from book work. Access to hands-on projects and a sense of freedom and agency is limited at best.... (read more)

Self-study ideas for micro-projects in "abstract" subjects?

Always enjoying your thoughts. Thanks chemslug. My expectation is that there are more safe and tractable micro-projects out there than the average student takes for granted. But I am also raising these questions to confirm a suspicion I have: that despite our love for the idea of "learning by doing," there are many disciplines where a long slog of paper-based learning, punctuated by a few carefully regulated experiments, has to precede any kind of creative or independent hands-on activity.

LessWrong's steeped in "move fast and break things" Silicon Valley c... (read more)

8chemslug6dWhat an "aspiring chemist" should do depends a lot on age and where they are in the educational process. For children below high-school age, I think there are lots of great experiments you can do to illustrate principles of chemistry. Lack of originality isn't a bug there, it's a feature. In high school, if you think you like science, take chemistry! There should be a lab component in most schools, so you can at least get a flavor for what working with chemicals is like. Access to equipment like this is an underrated component of the educational system. For college students, all the entry-level courses (general, organic, inorganic) are likely to have lab components. There are a few programs that separate the lab courses from the lectures, but they're the exception. The experiments won't be cutting-edge, but rather are designed to give students an understanding of chemical principles and what it's like to work with chemicals in a research setting. Where things get really interesting from a lab perspective is if you can convince a professor to let you into their research lab as an undergraduate assistant. It's helpful if you're at a research university rather than a small liberal arts college because the lab facilities will be more conducive to cutting-edge research, and there will be grad students and postdocs who relish the opportunity to teach a curious undergrad how to do chemistry. You likely won't be designing your own project, but will have the opportunity to use modern equipment to do novel research under supervision. My undergraduate research experience was formative and a huge reason why I do what I do today. I was also lucky enough to get to pay it forward and mentor undergraduates when I was a grad student and a postdoc. Graduate school in chemistry has a (somewhat deserved) reputation as a potentially miserable time. You get lots of great experience and training, but the hours are brutal and the rewards can be sparse. Synthesis in particular is a field
Self-study ideas for micro-projects in "abstract" subjects?

These are great ideas, thanks! I like both the specific suggestions and the idea of contacting societies popularizing the subjects. I hadn't thought of that idea.

Agreed that there are real issues with doing "interesting" chemistry at home. I do think that this is one area where the idea that students can motivate themselves by finding an "applied" project to work on might not be the best advice.

Self-study ideas for micro-projects in "abstract" subjects?

Here's another way of stating the exercise:

  1. Take a subject you're familiar with, or ideally are an expert in.
  2. Pick a sub-sub skill out of an introductory textbook on the topic. For example, if your subject is biochemistry, you might pick "RNA aptamers." If the subject is differential equations, you might pick "separable equations."
  3. Propose a micro-project where a student who had an introductory-level grasp of that topic would find it directly relevant. This micro-project should ideally be safe, cheap, legal/ethical, tractable for a self-studying introductory
... (read more)
Search-in-Territory vs Search-in-Map

Those are nice examples/test cases!

Here's what I think is the right way to understand what's going on in the phone case. Let's say you're looking for an ice cream stand in a park.

Your brain takes input from the phone and your eyeballs. It synthesizes them, along with memories and other sense data, into a prediction about where you should walk and what you should look at in order to find the ice cream stand. Based on that mental synthesis, it sends outputs to your body, causing you to walk/read/look around.

In this conception, there's ultimately only "search... (read more)

How do you establish a comfort zone in your studies?

Makes perfect sense! Going into scientific research, there's often kind of a chicken-and-egg problem. You know that knowing biochemistry, differential equations, immunology etc. would be useful somehow, eventually, if you became familiar with these subjects. You do have a general interest in these topics, since they're directly relevant to becoming, say, a biologist. But most of your learning along the way won't be directly related to a project that directly has real, applied value to your own life or the lives of others. That comes later. So figuring out ... (read more)

2nim6dYou highlight a difference that relates to why I don't feel like I do my best work in academia. I think an overarching project -- "I want to learn enough biology to cure cancer" or "I want to learn enough electrical engineering to design audio equipment" or even "I want to learn enough marketable skills to make a truckload of money" can turn academics into project-aligned work. However, looking for one's personal project or motivation for being in academia and finding only "well I guess people praised me when I said I wanted to be a scientist" or similarly uncompelling motives can be dangerously demotivating.
How do you establish a comfort zone in your studies?

Agreed. I think of this as the problem of "source selection." C.f. The Best Textbooks on Every Subject if you haven't checked that out, though I don't know if I agree with the recommendations or this anecdotal approach to the problem.

How do you establish a comfort zone in your studies?

Good thoughts. I agree that having a convenient practical application is very nice. Programming is lovely, because even a complete novice can make things that feel interesting to them with just a few pieces of basic knowledge.

By contrast, there's a fair bit of precursor knowledge required to figure out how to apply, say, differential equations to a biological modeling problem. Even though chemistry is in theory practical, the danger, regulation, and expense of setting up a laboratory to mess around in makes practical projects a less appealing way to learn ... (read more)

3Selueen9dI understand your point, and I for the most part agree. It is important to understand the basics. What I was trying to say is.. If you did not get the basics from your first attempt to learn those, maybe try to approach them differently. Look for a different textbook, ask someone who is not your current teacher, maybe look for popular explanation (if you are compltetly lost), or for more technical one (if original was not detailed enough), etc etc. Try to learn the basics, but switch the approaches if you are stuck. I feel like it might help with motivation too, as it should be more exciting than plain repetition.
Search-in-Territory vs Search-in-Map

It seems to me like search in territory (SIT) and search in map (SIM) are matters of degree, not kind. So they can potentially be quantified. They also have to do with transduction from one form of information to another.

For example, with the SIT example, you’re transducing information from scale balance and rock position into and out of brain states. With the SIM example, you transducer information from your brain, into a pre-designed spreadsheet, then from scale balance and rock position into your brain, into a spreadsheet, and then back to rock position.

It doesn’t seem like there’s a hard distinction between the two from that perspective? Not sure.

4philh8dI'm not sure if these are examples of the thing you're talking about or something else, but: Consider a missile that's guided by GPS until it reaches its rough target location, then uses sensors to locate the target precisely. (Though arguably this is simply "SIM followed by SIT".) Or consider when I do something similar myself. I use the map on my phone screen to guide me to roughly where I want to be, and then I use my eyes to guide me to exactly where I want to be. And I don't just switch from SIM to SIT; I keep checking with both, in case e.g. I miss it and go too far.
Alcohol, health, and the ruthless logic of the Asian flush

From Wikipedia: "Disulfiram does not reduce alcohol cravings, so a major problem associated with this drug is extremely poor compliance. Methods to improve compliance include subdermal implants, which release the drug continuously over a period of up to 12 weeks, and supervised administration practices, for example, having the drug regularly administered by one's spouse."

My guess is that for a strategy like this to work better, you'd need the pain to come right away, strong enough to build an immediate association between alcohol and suffering. Instead, "a... (read more)

What are your personal (public) Hamming questions?

Hi weathersystems, I like this idea. I have a few reactions to it.

First, it sounds like to be a success, you just need to find one other person to collaborate with. If you can find that person, go for it!

Secondly, if your goal is to get more people interested and more questions submitted, I think it's worth taking more time to have individual conversations with specific people about topics you think they'd be interested in collaborating on based on their post history. Sussing out their level of interest, availability, and what sort of collaborative partner... (read more)

For Better Commenting, Take an Oath of Reply.

That's a good point. I picked the word "oath" intuitively, and I can try to articulate why.

First, "policy" feels more detached and state-ish than I wanted. I wanted a word that conveyed some emotional depth and a spark of human connection.

Also, "policy" has the implication of being explicit in its details, like a law. By contrast, an oath is about building credibility without being specific about what actions to take. Here's a sample from the Hippocratic Oath:

  • "I swear to fulfill, to the best of my ability and judgment, this covenant: I will respect the har
... (read more)
2jimrandomh16dIn most cases I think the correct phrase would be "nearly unlimited". It unpacks to: the set of circumstances in which a limit would be reached, is nearly empty.
AllAmericanBreakfast's Shortform

Pet peeve: the phrase "nearly infinite."

3JBlack16dWould you prefer "for nearly all purposes, any bounds there might be are irrelevant"?
Why don't long running conversations happen on LessWrong?

Adam Zerner’s recent question is a good example of a “this isn’t working” followed by analysis of the causes. Emotion is often a motivator for meta conversation in offline relationships. In online settings like LW, we might tend to start by imagining what sort of interactions we think are possible and missing, and then inferring the reason for their absence and experimenting with solutions.

Keeping those analyses general and succinct seems like the right way to go. I wouldn’t want to have a public discussion about my feelings about a specific other blogger. But trying to find explanations for general behavior online seems like world modeling, not navel gazing.

3Willa1dI agree re Adam's question being a good example of the case, and not at all navel gazing. My worry about navel gazing is motivated by some personal baggage, anxiety about what's worth spending time on, and some anecdotal experiences where it seemed like metaconversations were not effective in doing anything and took a ton of time and energy, then nothing happened from them. Focusing on "...trying to find explanations for general behavior online..." definitely is helpful world modeling that can provide useful insights and instigate changes in behavior in response to those insights. I like how you only used the word meta conversation in your comment a single time and tabooed it otherwise, instead offering more specific and actionable commentary / insights about the issues of online discussions and how to align those with what people want out of the site + discussions. Taking an Oath of Reply [] like you wrote is a good idea, and I'll start doing that for my own posts or on discussions / comment threads too. "In online settings like LW, we might tend to start by imagining what sort of interactions we think are possible and missing, and then inferring the reason for their absence and experimenting with solutions." This would make a really good question, if it hasn't already been asked recently I'll go make that post and reference this discussion, post, and other relevant things.
What are your personal (public) Hamming questions?

Here's a description of the Hamming question.

Mathematician Richard Hamming used to ask scientists in other fields "What are the most important problems in your field?" partly so he could troll them by asking "Why aren't you working on them?" and partly because getting asked this question is really useful for focusing people's attention on what matters.

I'm not trying to troll anybody. The three questions at the top of the OP are what you'd answer to articulate your Hamming problem, as you've done here. What is the most important problem you face in your lif... (read more)

I grew up in poverty, so I've been a client of the Public System from birth. What this means is that I've been coordinating with all kinds of professionals to attempt to meet my needs for a long time. Like most domains/games/systems, the longer you are involved in/with them, the more you understand the limitations and problems of them. This is the case for me in trying to get out of poverty and homelessness. 

So while I can appreciate the concern for my safety, I can assure you I've been connecting with health professionals and their support systems fo... (read more)

What are your personal (public) Hamming questions?

Great, do you have time this week for a ~30 minute chat? :)

1Randomized, Controlled17dProbably! Do you use calendarly or something similar? Which timezone are you in?
For Better Commenting, Take an Oath of Reply.

I like the idea of a standard LW OoR, and maybe there could be several. Optimally it would be as convenient as possible to include one.

For Better Commenting, Take an Oath of Reply.

The oath of reply is meant as a prior-increaser for receiving a reply specifically for an audience aiming for good faith dialog. So any assumptions or modifications needed to specify the sort of conversation you’re looking to have are appropriate. Much of this needs to be implicit, as we’re all busy people and don’t need 10 pages of legal boilerplate for this stuff.

For Better Commenting, Take an Oath of Reply.

There are ways to do a “message received, thanks” in a few short words. Emojis are annoying in formal settings and convey much less than even just a single short sentence.

How concerned are you about LW reputation management?

Primarily real reputation, but interested in both aspects!

Core Pathways of Aging

Interesting! Can you give me some background on where those ideas come from? I haven’t specifically studied cancer biology yet so I don’t know if this is something an intro textbook on the subject would cover, or whether they’re to some extent your original ideas?

What I wrote isn't very original. 

The fact that immune function is worse in older people is standard knowledge and johnswentworth wrote a post about the thesis that the thymus might be a central factor here. 

The fact that cancers have to mutate to activate telomerase production to be able to constantly replicate seems to me like a cancer 101 thing. There might be some cancers that happen in stem cells that actually produce telomerase naturally but it's necessary for normal cells. 

A decade ago, cancer vaccines were targeting single targets an... (read more)

Why don't long running conversations happen on LessWrong?

Ah, what I mean is that if you specifically ask a person to discuss a topic, or bring it up clearly showing you want to discuss it with them, they’re likely to take you up on it in an organic conversation. Also, if you specifically ask a friend to give you feedback on a piece of writing, making it clear it would mean a lot to you, I wouldn’t be surprised if they do. Most people I know seem happy to help.

But yeah, just sending a friend a link to a post isn’t going to increase the response rate too much.

2adamzerner17dGotcha, I agree.
Why don't long running conversations happen on LessWrong?

Not the clearest language, sorry. Slightly more specifically, I mean that a lot of the conversation is either about our own conversations, conversational dynamics generally, or conversation dynamics between ourselves and family/friends or between people we know and others.

I don’t have a confident number to put on it. Between 10-50% perhaps? Which feels huge to me.

4adamzerner17dI see, that makes sense. That 10-50% range does seem huge to me also, but I think it's cool :)
For Better Commenting, Take an Oath of Reply.

Thanks for that great response, very in line with my own thinking!

4philh15dSo, this seems like a fine policy. But calling it an "oath of reply" feels like it waters down the word "oath" in a way I dislike. (cf. the people who've taken the GWWC pledge, and said that if in future they think it's not a good idea, they'll just stop doing it.) Especially when the stuff you've said should be a universal understanding (around self care, rudeness, circumstances changing) is left implicit. That stuff won't be universally understood. As a simple change that I personally would consider an improvement, I'd call it a "reply policy". A few words making it clear that this not absolute might be good too. Perhaps, for the one you left on this post: (Possible downside: readers might take that to be weaker than you consider it.)
For Better Commenting, Take an Oath of Reply.

gjm expresses it well! I think you’re right that some comments are in a special category of clearly not needing a reply. Maybe there’s a way to briefly acknowledge that the comment has been seen and considered, however.

4adamzerner18dEmoji replies come to mind. Although 1) I feel like there's something refreshing about them not being a thing on LessWrong and 2) I think it can be assumed that comments have been seen and considered.
What are your personal (public) Hamming questions?

So is your personal Hamming problem getting into a better environment for study? If so, what would be your first step? Are you working on it? And if not, why not?

You're clearly in a difficult situation, so know that the reason I ask those questions is because that's the theme of this discussion. There is no implicit criticism here. This is just a space to explore these questions, whatever the problems each individual person is facing.

6Josh Smith-Brennan18dMaybe it would help if I understood a little better what a 'Hamming Problem' was. It looks a bit to me like it has to do with issues of Communication Complexity (the Gap-Hamming Problem []?), so I'm assuming in the context of this post a Hamming Problem is one concerning inadequate understanding of (and so inadequate problem solving approaches to) a personal issue due to issues of complexity that can potentially be overcome by breaking the more complicated problem down into smaller, easier to tackle tasks? Although from the other posts there seems to be some consideration of adjacent or tertiary problems in relation to a Hamming Problem too. Overall, my life goal currently is to find some happiness and some long term financial and social stability. How that comes about is up in the air, but what I've been attempting to do is to take my experience and use it in conjunction with UX research/design skills to find some solutions to my problems. I was hoping to apply that work towards other issues besides personal ones, but circumstances caused me to start having to analyze the Public Health/Assistance system in the US from a client perspective. As a result, I started to develop a catalog of concerning issues - and began trying to design some countering solutions - with the Public System and how it interfaces with the Private System in order to create some sort of a social safety net. Turns out, at a certain point, the support these systems offer change dramatically, and the fall into homelessness is pretty quick, pretty complete, and pretty devastating, depending on where you started in life. So while trying to solve my own issues, I ended up attempting to understand the entire system, and some of the Systemic Biases and resulting problems. I would dearly love to separate out my issues from those of the 'system' so that I could work only on my personal problems to a satisfactory resolution, but because of the systemic
Why don't long running conversations happen on LessWrong?

It also makes me think about an idea I've always had: Blog Buddies. Alice is a blogger who is interested in having a lot of discussion and constructive criticism on her posts. So is Bob. And Carol. And Dave. They can all get together, form a group, and read + critique each others posts. Seems like it scratches an itch. I've thought about it as a startup idea, but I could see it making sense more narrowly on LessWrong.

Yeah, this sounds like a book club, except where the participants are also the authors. I'll bet that there are some groups of authors who do... (read more)

9Willa18dWhat metaconversations have you experienced in your closest relationships and/or elsewhere that you think LessWrong would benefit from discussing? How do you turn the metaconversations into actionable, implemented, and solved advice? Rather than such conversations well...becoming very navel gazing and meta circle jerk-y? I imagine it's much easier making such conversations bear fruit and pay rent in the context of close personal relationships because there should be a more visceral "this isn't working" type of feeling fairly immediately, yes? Whereas in an online, usually loosely connected social environment I imagine the visceral and immediate feelings of "this doesn't work" probably don't arise as much, or at least not quickly. This is me speculating, what do you think? I'll second the "posting lots of questions" being more catalyzing and accessible feeling than regular posts. I still don't comment too much nor write as much as I'd like, but whenever I see someone post a question post that seems to generate more discussion with people who's names I don't recognize than other types of posts. And seem more accessible. Anecdote: I comment way more on metaconversation and community norms / culture types of posts than I do ideas focused and other kinds of posts. Not 100% certain why, might have to do with issues concerning people, group and community norms, culture, etc. feeling more accessible and interesting to me whereas pure ideas are less interesting to me usually. I like people a lot and seem to get more interested in something based on the impact that thing has on people. Probably is why I find X risks, AI Safety, and other such things to be very important and good and have read a good bit about such things, but whenever I try to dig into the weeds of the ideas and grok the technical idea aspects of those things...well that's a lot less interesting to me. Anyway, this was an aside. Helpful to me though.
4adamzerner18dI like that analogy. Yeah I like that idea. Huh, interesting. Drawing that connection to your social life makes it feel intuitive to me that we should be doing a lot more of it on LessWrong. Seems like a good topic for a follow up post. Just curious, but if you don't mind sharing, to what extent is this an exaggeration?
Why don't long running conversations happen on LessWrong?

I'm having trouble understanding what you mean here. Would you mind explaining this differently?

You're right, that was convoluted :)

Having a prior that your conversation partner will reply motivates commenting. When both partners have that prior, they'll continue their conversation, as we're doing here. Figuring out how to generate that prior in both parties is the important part. So how do we generate that prior?

One way is trust/personal connection. If you're having a conversation with a friend, you can usually expect they'll respond when you bring up a n... (read more)

4adamzerner18dThis makes perfect sense now, thanks! I actually don't get a strong sense that this is true. The handful of people on LessWrong who I'm friends with, I expected that they're slightly more likely to respond to my posts, but not by too much. I expect it to be mostly about whether they are interested in the post and have something to say.
2adamzerner19dCool, I'll check it out!
Why don't long running conversations happen on LessWrong?

This is partly why I’ve shifted in my posts on scholarship to just building on my own thoughts over a long time.


Hm, previously I had been assuming that this wouldn't work because readership isn't necessarily consistent, and because even if you do get the same readers they'd have lost the context of the previous posts.

Oh, to be clear, this is a response to having to some extent given up on hoping for deep conversations from readership. Instead, I just try to structure my research such that I have a feeling of responding to and building on my own line o... (read more)

5adamzerner19dAh, I see. I do see the value in this and might give it a shot myself. It also makes me think about an idea I've always had: Blog Buddies. Alice is a blogger who is interested in having a lot of discussion and constructive criticism on her posts. So is Bob. And Carol. And Dave. They can all get together, form a group, and read + critique each others posts. Seems like it scratches an itch. I've thought about it as a startup idea, but I could see it making sense more narrowly on LessWrong. Yeah that makes sense. I've had some similar behind the scenes attention happen before. This makes me think of something Paul Graham says. That it's better to have a small amount of users who really love you than to have a large amount of users kinda like you. He's referring to startups, but perhaps it applies to writing as well. I'm having trouble understanding what you mean here. Would you mind explaining this differently? Yeah. I was going to link to a Beeminder goal for the pledge I made on this post, but I agree with you and didn't want to start off with such a heavy handed convention. Cool, sounds good!
What are your personal (public) Hamming questions?

When I try to choose between any of the plethora of options I could put time towards, they all just feel sort of "flat" and difficult to differentiate. I think this is also tied up with often finding it hard to be strategic when I settle down infront of my computer -- a sense of brain fog often settles in.

tl;dr: I think in some sense organization is my hamming problem

Any chance you'd be interested in a small "feel it out" collaboration on my project for studying how to study? It sounds like you're studying, and also trying to decide what to study, both of ... (read more)

3Randomized, Controlled19dFor sure, that would be fun : )
What are your personal (public) Hamming questions?

That sounds like an amazing project! Do you think there's a chance that your stuck-ness is caused by you thinking your job and kids are more important than this video project? Maybe your Hamming project is being a parent and doing your day job, in which case you actually are working on your most important things?

What are your personal (public) Hamming questions?

Sounds like you want to create a software library for gameification (both implementing the features and increasing their addictiveness)?

It sort of sounds like you want to take the symmetric weapon of addictiveness, which the forces of doom have perfected and the forces of hope have neglected, and make it accessible to the good guys?

When I decided to pursue an EA career, I started taking classes and committed to grad school. This fall I'm moving to Michigan for an MS in bioengineering. Obviously that's not concretely relevant to your project.

However, in the... (read more)

What are your personal (public) Hamming questions?

The timescales to get a full feedback loop (think of an idea, try it out, analyze the results) often look like weeks to run an experiment, whereas in the digital world your code will compile in seconds.

I think this is superficially compelling. After all, if you read a biology method, often a single step will take hours, days, even weeks. By contrast, of course, a single line of code executes in next to no time.

However, it's not actually the right way to think about it. After all, there's no reason that we would compare a single physical action, or a single... (read more)

Why don't long running conversations happen on LessWrong?

This is partly why I’ve shifted in my posts on scholarship to just building on my own thoughts over a long time. Having a readership, even if shallow, is motivating. I do occasionally get useful comments, sometimes via PM. I think reaching out to others for zoom or other back and forth goes a long way to establishing the trust needed to generate even a 45 minute conversation.

With commenting, I feel at each stage a concern that I won’t be responded to. I imagine others feel the same way. And this frequently turns out to be the case! Many times I’ve taken a ... (read more)

4adamzerner19dHm, previously I had been assuming that this wouldn't work because readership isn't necessarily consistent, and because even if you do get the same readers they'd have lost the context of the previous posts. But in thinking about it now, I'm shifting a small-moderate amount away from this belief. My model is different here. I don't get the sense that trust/personal connection is something people particularly look for as a prerequisite to having a long running conversation. Thanks for contributing a data point here. I've noticed this too and very much agree. I agree, but don't feel strongly enough to think that these things aren't worth experimenting with (the bar for experimenting is a lot lower). In particular, my model is that heavy handed incentives would be more important for the other party. Eg. if Alice knows that there are people who pledged real money, she'll feel more confident that they will keep their word and participate at a high level. On the other hand, if Alice only has a measly promise from the other people, she may think, "Meh, who knows whether they'll really keep their word." Hey, that's a great MVP! I'll try it for this post! (Done [] )
Demons in Imperfect Search

Another example might be democratic politics. Optimization is meant to produce a government and policies representing a majority view while protecting minority rights. Search is via voting, a procedure which is defined in a difficult-to-change constitution; politicians who are elected have an incentive to preserve the system that got them elected. Exploitation happens when actions that would better represent majority views and protect minority rights don’t necessarily get politicians elected. In fact, there are actions politicians can take to further decouple representation and rights-protection from voting.

Demons in Imperfect Search

Addiction might be another example. It starts with pursuing a feeling of relief. Search is imperfect, focusing on reward system responses in the brain rather than the feeling of relief originally sought. Drug makers and addicts focus on stimulating that reward center, rather than on creating/consuming drugs that might produce relief. Some actions that stimulate the reward system further decouple brain stimulus from relief, like self isolation or theft to get money for drugs.

4johnswentworth18dExcellent example. Your politics example is great too.
Two Definitions of Generalization

This article on “Generic Generalizations” seems potentially relevant?

4ozziegooen20dYes, thanks!
The Homunculus Problem

"We don't see the actual colors of objects. Instead, the brain adjusts colors for us, based on surrounding lighting cues, to approximate the surface pigmentation. In this example, it leads us astray, because what we are actually looking at is a false image made up of surface pigmentation (or illumination, if you're looking at this on a screen)."

Cultural concepts and the colloquial language of vision do not map very neatly onto the subjective experience of vision. Nor does the subjective experience of vision map neatly onto the wavelengths of light striking... (read more)

What are your personal (public) Hamming questions?

Thanks for the recommendation!

I love programming. Taught myself back in high school, and it's such a lovely self-reinforcing and immediately useful skill.

Now I'm back in school after 10 years working in music. This fall I'm attending an MS program in bioengineering, which I'm QUITE excited for. One of the things about studying biology is that, well, biology is amazing, but it's a lot harder to do even basic things, much less create anything original, at home. One of the coolest resources is, which has lots of bioinformatics practice problems. I haven't seriously looked into the biohacking community, but it's time.

3Austin Chen20dI deeply agree with your perspective on biology - there's so many cool things imaginable (curing mortality, augmenting health and mental capabilities) but progress feels hard to come by because of high barriers to doing basic work. The timescales to get a full feedback loop (think of an idea, try it out, analyze the results) often look like weeks to run an experiment, whereas in the digital world your code will compile in seconds. [] is really neat, though fundamentally I wouldn't have imagined bioinformatics/data processing to be the rate limiting step in advancing biology research... And btw for other readers: I realized that my earlier examples may have been construed as "everyone should be programming" - which is a view I might also endorse, but not my main point. My intent was closer to "in any field you're trying to study, you'll retain more knowledge and be more motivated by the process of creating something others value". That could look like: LessWrong posts summarizing state of the art research for a grade school audience, or creating a Discord for fellow students to exchange ideas.
What are your personal (public) Hamming questions?

I'll go first. I've been working for a year on a "studying how to study" project. There are a couple potentially high-impact experiments I could do.

  1. Buying or synthesizing modafinil (the synthesis is very easy). First step would probably be to buy some via an online pharmacy, since the synthesis would be time-consuming and expensive unless I do it in large quantities.
  2. Going on a 10-day study retreat as suggested by Razied. First step would be announcing to friends and family that I intend to take 10 days in August to try this.

I also need to figure out what i... (read more)

1weathersystems14dI think some question in this area would work well for this collaboration I'm proposing: [] If you add a question there and it gets picked I'd be happy to work on this with you.
6Austin Chen21dIn case having concrete steps on buying modafinil would help: I've had two good purchasing experiences with [] in the last few months. They take PayPal, no messiness with eg bitcoin needed. Re: studying, it sounds like you're still in school? I studied a lot in school and got top grades; same for many of my friends. None of us wish we "studied harder", looking back. Instead of focusing on studying for the sake of it, just go out and create something valuable! The periods I learned the most did not look like "studying more", they looked like "teaching for an undergrad CS class", "building mobile apps with a team of friends, and publishing them on the Google Play Store".
(Trying To) Study Textbooks Effectively: A Year of Experimentation

As a followup, it does seem like you could test this to some extent. Find out what textbook an upcoming year-long class uses. Take 10 days prior to the class for a retreat, during which you read the entire textbook. Then try to do the homework and exams with minimal review of the textbook, treating the class as a review of material you're already familiar with rather than a first brush with the content. Since textbook reading comprises the bulk of my studies, it seems possible that this would feel like a net time saving/deeper learning, but I'm not sure. Would be interesting to try it and see!

Load More